Sphingosine 1-phosphate (S1P) is a bioactive molecule that signals by activating S1P receptors (S1P1-5) to regulate cellular processes in systemic immunity and inflammation. Accumulating evidence indicates that ab- normalities of S1P signaling are associated with brain aging and with the pathophysiology of Alzheimer's dis- ease (AD). Fingolimod, a sphingosine analog used for the treatment of multiple sclerosis (MS), acts through S1P1 in lymphocytes to reduce their infiltration into the CNS and thereby provides therapeutic effects against subsequent neuroinflammation. Fingolimod crosses the blood brain barrier and emerging evidence suggests the direct neuroprotective effect of fingolimod on CNS cells. In experimental models of AD fingolimod appears to reduce the production and the neurotoxicity of A? peptide and promote neuroprotection of microglia and neurons. We have reported that in 5xFAD mice, a transgenic model of AD, oral fingolimod treatment decreases the activation of microglia and reactive astrocytes, decreases A? levels, and increases hippocampal neuro- genesis. Our preliminary data show that most of the neuroprotective effects of fingolimod in 5xFAD mice occur at a low dose with major effects on neuroinflammatory markers. We hypothesize that aging alters the S1P sig- naling system in the brain and drives the proinflammatory activation of astrocytes and microglia that is acceler- ated by the buildup of A? and that treatment with S1P modulators will interfere with this process and may affect age- and AD-related neuropathology and behavioral deficits. To test this hypothesis we will use two different transgenic mouse models of AD (5xFAD and PSAPP) that accumulate A? at different rates such that similar amounts of A? are deposited in the brain at different ages. The study includes 3 aims. The first two aims are mechanistic studies to determine the effect of age and AD-like pathology on the S1P system in wild type mice and in the mouse models of AD (aim 1) and to determine the effects of S1P receptor modulators on neuroin- flammation in aging and AD mouse models and on AD-related neuropathology and cognitive function in these models (aim 2). The latter studies will include a prevention arm by treating the mice before the emergence of AD-like pathology until old age (1-18 months), and an advanced disease arm by treating the mice with well- established pathology (15-18 months). The outcome measures will include cognitive behavior, amyloidosis, ac- tivation of astrocytes and microglia, S1P system, neurotrophin signaling, e.g. BDNF/TrkB, cytokines, synaptic- glial- and apoptotic markers, and small molecules determined by magnetic resonance spectroscopy (MRS) that reflect neural metabolism, excitotoxic, oxidative, and osmotic stress, as well as membrane integrity. Machine learning tools will be employed to integrate these data sets. The third aim will be the translational arm of the study to determine the effects of S1P modulation on indices of neuronal and glial function in aging and in AD mouse models using noninvasive techniques including in vivo magnetic resonance spectroscopy imaging and spectroscopy (MRI/MRS) and positron emission tomography.